Identification and prioritization of optimum capacity solutions in a telecommunications network

Abstract

Systems and methods that use historical data comprising capacity gain solutions and their associated gains at various locations to train a machine learning model. The trained machine learning model, upon receiving a new location (e.g., latitude and longitude coordinates), recommends the top n (e.g., the top 3) solutions that should be deployed at the new location to improve telecommunications network performance. The machine learning model uses clustering techniques to perform the recommendations.

The invention claimed is: 
     
       1. An apparatus to identify network improvement solutions deployable at one or more locations in a wireless telecommunications network, the apparatus comprising:
 at least one data processor; and 
 at least one memory, communicatively coupled to the at least one data processor, and storing instructions executable by the at least one data processor, wherein the instructions comprise:
 using historical data to train a machine learning model,
 wherein the historical data includes capacity gain solutions, and associated gains achieved based on the solutions, and 
 wherein the capacity gain solutions were previously implemented at various geographic locations associated with the wireless telecommunications network; 
 
 upon receiving a new geographic location, generating, with the trained machine learning model, recommended solutions deployable at the new location to improve performance of the wireless telecommunications network at the new location,
 wherein the machine learning model uses clustering techniques to perform the recommendations. 
 
 
 
     
     
       2. The apparatus of  claim 1 , wherein the instructions further comprise:
 building or accessing a dataset comprising information of previously deployed capacity improvement solutions within the wireless telecommunications network, wherein the information includes data related to gain, cost, location, duration, and solution type. 
 
     
     
       3. The apparatus of  claim 2 , wherein the clustering includes clustering on the information in the dataset for each recommended solution, and categorizing each recommended solution within a geographic area. 
     
     
       4. The apparatus of  claim 1 , wherein the clustering includes ranking each recommended solution in each cluster based on one or more of: spectrum, duration, area, or cost. 
     
     
       5. The apparatus of  claim 1 , wherein the instructions further comprise:
 receiving a user input of a new geographic location; 
 finding a nearest cluster or recommended solutions, and 
 outputting a top n number of recommended solutions for the nearest cluster. 
 
     
     
       6. The apparatus of  claim 1 , further comprising receiving input from a user clicking on a location on a displayed map to identify the new geographic location. 
     
     
       7. At least one computer-readable medium, excluding transitory signals and carrying instructions, which when executed by a data processor, performs operations for a wireless telecommunications network, the operations comprising:
 using historical data to train a machine learning model,
 wherein the historical data includes capacity gain solutions, and gains achieved based on the solutions, and 
 wherein the capacity gain solutions were previously implemented at multiple geographic locations associated with the wireless telecommunications network; and, 
 
 upon receiving a new geographic location, generating, with the trained machine learning model, recommended solutions deployable at the new location to improve performance of a wireless telecommunications network at the new location,
 wherein the machine learning model uses clustering techniques to perform the recommendations. 
 
 
     
     
       8. The at least one computer-readable medium of  claim 7 , wherein the instructions further comprise:
 building or accessing a dataset comprising information of previously deployed capacity improvement solutions within the wireless telecommunications network, wherein the information includes data related to at least three of: gain, cost, location, duration, or solution type. 
 
     
     
       9. The at least one computer-readable medium of  claim 8 , wherein the clustering includes clustering on the information in the dataset for each recommended solution, and categorizing each recommended solution within a geographic area. 
     
     
       10. The at least one computer-readable medium of  claim 7 , wherein the clustering includes ranking each recommended solution in each cluster based on one or more of: spectrum, duration, area, or cost. 
     
     
       11. The at least one computer-readable medium of  claim 7 , wherein the instructions further comprise:
 receiving a user input of the new geographic location; 
 finding a nearest cluster or recommended solutions, and 
 outputting a top n number of recommended solutions for the nearest cluster. 
 
     
     
       12. The at least one computer-readable medium of  claim 7 , further comprising receiving input from a user clicking on a location on a displayed map. 
     
     
       13. At least one computer-readable medium, excluding transitory signals and carrying instructions, which when executed by a data processor, performs operations, comprising:
 receiving data related to existing capacity planning solutions deployed at multiple existing locations; 
 training a machine learning model based on the existing capacity planning solutions data,
 wherein training the machine learning model includes creating data clusters based on the existing capacity planning solutions data; 
 
 receiving latitude and longitude coordinates of a new geographic location; and 
 applying classification techniques to determine optimum capacity planning solutions capable of being deployed at the new geographic location using the created data clusters, to thereby efficiently and economically identify solutions and locations to expand capacity of cell sites at the new location within a wireless telecommunications network and improve telecommunications network performance. 
 
     
     
       14. The at least one computer-readable medium of  claim 13 ,
 wherein receiving data includes receive historical data having capacity gain solutions and associated gains at the multiple existing locations; and 
 wherein the applying includes recommending, using the trained machine learning model, a top n solutions to be deployed at the new location. 
 
     
     
       15. The at least one computer-readable medium of  claim 13 , further comprising:
 building or accessing a dataset comprising the existing capacity planning solutions data, wherein the existing capacity planning solutions data includes at least two of: gain, cost, location, duration, or solution type. 
 
     
     
       16. The at least one computer-readable medium of  claim 13 , further comprising:
 clustering the existing capacity planning solutions data per market for each of multiple solutions to categorize solutions within a geographic area. 
 
     
     
       17. The at least one computer-readable medium of  claim 13 , further comprising:
 ranking the solutions in each data cluster based on: spectrum, duration, latitude/longitude area, cost, or any combination thereof. 
 
     
     
       18. The at least one computer-readable medium of  claim 13 , further comprising:
 receiving user input of the new location via entering of the latitude or longitude or by clicking on a location on a map. 
 
     
     
       19. The at least one computer-readable medium of  claim 13 , wherein the applying includes finding a nearest cluster and showing a top n solutions in the nearest cluster.